Autopsies of soldiers killed by mustard gas in World War I indicated that sulfur mustard has a disproportionate effect on rapidly dividing cells and suggested that sulfur mustard compounds might have antitumor effects. Indeed, early researchers attempted to treat cancer by direct injection of sulfur mustard into tumors. This research was limited by the extreme toxicity of sulfur mustard compounds and nitrogen mustard analogs, such as mechlorethamine, were investigated as less toxic alternatives.

In general mustard compounds exert their cytotoxic effects by alkylating DNA, such as at the N-7 position of a guanine residue. The mechanism of alkylation by mustard compounds is illustrated in Scheme 1. With reference to Scheme 1, mustard compounds have an internal nucleophile that assists in chloride displacement by, as shown for the case of mechlorethamine, forming an aziridinium intermediate. Because mechlorethamine has two leaving groups, the nucleophilic substitution mechanism depicted in Scheme 1 can be repeated resulting in a DNA or protein-DNA crosslink.

Mechlorethamine is extremely reactive and as a result is non-selective. Thousands of alkylating agents have been designed and prepared using mechlorethamine as a model. However, few of these compounds have demonstrated sufficient therapeutic superiority to mechlorethamine to warrant clinical trials.
Because of the lack of selectivity of most methchlorethamine analogs, prodrugs, such as phosphoramide compounds, which can be activated by the high concentration of phosphoramidases present in neoplastic cells, have been investigated. Two phosphoramide alkylating agents, cyclophosphamide (CPA) and the isomeric compound Ifosfamide (Ifos) have proved to be particularly effective.

The metabolic pathway of CPA is similar to that of Ifos (the metabolism of Ifos is illustrated in FIG. 1) and thus the two compounds share common drawbacks. Perhaps most important is their dose limiting toxicity due to hemorrhagic cystitis. The hemorrhagic cystitis is believed to be induced by the production of acrolein during the activation of both CPA and Ifos.
Acrolein is an active electrophile that reacts with thiols under physiological conditions, which may be responsible for its liver toxicity in the form of glutathione depletion. Finally, acrolein has been demonstrated to be a teratogen and a potent mutagen, and this may be responsible for the link between CPA treatment and serious side effects, such as bladder carcinoma and other malignancies.
With reference to FIG. 1, isophosphoramide mustard (IPM) is a common metabolite of CPA and Ifos. IPM is thought to be responsible for at least a portion of the anti-tumor activity exhibited by CPA and Ifos. Efforts to use IPM as an anticancer agent directly have been unsuccessful due in part to the compound's instability. IPM has been synthesized and preliminary biological evaluations of the compound have been conducted, but unfortunately IPM is too unstable to be isolated and used for human treatment.